Factors associated with atypical speech representation in children with intractable epilepsy

Abstract

The purpose of this study was to investigate the relationship between hemispheric speech dominance and several demographic and seizure variables in a large sample of children (N=75) who underwent the intracarotid amobarbital procedure (IAP) prior to surgery. The findings were similar to results previously reported in adults and children, suggesting that variables, such as age at seizure onset, laterality of seizure focus, location of seizure focus and handedness are related to the presentation of atypical speech representation in children with epilepsy. The impact of multiple variables in determining the risk for atypical speech dominance was also assessed. The results demonstrated an increased probability of atypical speech representation in children with a greater number of risk variables, suggesting the need for further investigations in this area.

Introduction

The usefulness of the intracarotid amobarbital procedure (IAP), sometimes referred to as the Wada test (Wada & Rasmussen, 1960), in identifying hemispheric language dominance in adults is well documented. The availability of these data has served to increase our understanding of the relationship between hemispheric speech representation and several demographic, neurological and psychological factors. Although some studies have included individuals with childhood seizure onset in their samples, the number of these individuals is often small, and may be combined in the same group as individuals with seizure onset in adulthood. IAP has become a common procedure with pediatric patients prior to surgical intervention for epilepsy (Szabo & Wyllie, 1993). The ability to accurately identify the locus of language functioning in the brain is crucial to mapping a surgical approach that will spare language skills to the greatest extent. The increasing number of cases of children undergoing IAP provides an opportunity to further examine the relationship between language dominance, and demographic/psychological variables in samples with early onset of seizures. This study aimed to add to the existing understanding of these relationships by investigating variables associated with atypical speech representation in individuals with childhood onset of seizure disorder.

In order to identify factors that may be associated with the pattern of atypical speech representation in children with epilepsy, variables were chosen based on findings from previous studies of speech representation. Several important patterns were identified, and are discussed according to the variables described below.

Several studies involving pediatric or mixed (adult and child) samples have suggested that the development of atypical speech representation is more likely when there is an early (versus late) lesion (Duchowny et al., 1996; Isaacs, Christie, Vargha-Khadem, & Mishkin, 1996; Mueller et al., 1999; Rey, Dellatolas, Bancaud, & Talairach, 1988; Springer et al., 1999). The methodology used to determine speech dominance and the definition of “early” versus “late” onset has varied across studies, and there has been no consensus as to the age at which the development of atypical speech is likely to decline. Thus, the goal of this study was to (1) examine language dominance as a function of the age at seizure onset, and (2) investigate the patterns of language dominance in “early” versus “late” seizure onset groups defined in terms of a cut-off point (i.e. a window during which atypical speech representation is more likely to develop).

Since the left hemisphere is dominant for language in the majority of right- and left-handers, the expectations for speech reorganization are intuitively greater following left hemisphere injury (Knecht et al., 2000). However, studies that have examined the relationship of laterality and speech dominance have yielded mixed results. Several studies have supported the relationship between left hemisphere seizure activity and hemispheric language reorganization (Rausch & Walsh, 1984, Rasmussen & Milner, 1977, Westerveld et al., 1994, Williams & Rausch, 1992). Rey et al. (1988) found no relationship between side of epileptic focus (left or right hemisphere) and speech dominance overall. Woods, Dodrill, and Ojemann (1988) reported that, in adult patients with extratemporal lesions, left-sided lesions were more likely to be associated with atypical speech representation than right-sided lesions. Hinz, Berger, Ojemann, and Dodrill (1994) reported a strong correlation between right-sided hemiparesis (i.e. early left hemisphere injury) and atypical speech representation in their sample of children who underwent IAP. In contrast, DeVos, Wyllie, Geckler, Kotagal, and Comair (1995) reported 10 cases of early low-grade left frontal or left temporal tumours that did not result in transfer of language dominance as measured by IAP. Similarly, Duchowny et al.’s (1996) results from subdural grid recordings in children supported the hypothesis that developmental pathologies may not result in transfer of language dominance unlike what has been reported following early acquired lesions. Of the patients they studied, there were 12 in whom language cortex was detected in proximity to or overlapping with the epileptic region in the left hemisphere.

Few studies have specifically examined the relationship between lesion localization and the incidence of atypical speech representation. Woods et al. (1988) reported that abnormal extratemporal findings on neuroimaging in adults were associated with an increased incidence of atypical speech dominance. This relationship could not be examined in patients with focal temporal lesions, however, due to small sample size. DeVos et al. (1995) also reported that neither early low-grade frontal nor temporal tumours results in a shift of language dominance to the contralateral hemisphere. According to these authors, language centres presumably developed in areas of the left hemisphere not occupied by the tumour as the children did not generally suffer impairments in language functioning following tumour resection.

Studies of the relationship between handedness and language dominance have been remarkably consistent in demonstrating that individuals who are left-handed are more likely to have atypical speech representation than people who are right-hand dominant. This finding has been reported in healthy adult volunteers (Van der Kallen et al., 1998) as well as adults (Milner, Branch, & Rasmussen, 1964; Rey et al., 1988, Woods et al., 1988) and children with brain lesions (Hinz et al., 1994, Milner et al., 1964; Vargha-Khadem, O’Gorman, & Watters, 1985). Woods et al. (1988) were careful to point out that, “this apparent association (between handedness and language dominance) may be due to a true association between handedness and speech representation, or it may result at least in part from an association of left handedness with brain lesions that are more likely to result in atypical speech representation” (p. 515). Milner et al. (1964) found that “when left-handedness was secondary to early damage to the left hemisphere, right-sided speech representation was more common (than in left-handers without evidence of an early left hemisphere lesion)” (p. 213).

Strauss, Wada, and Goldwater (1992) reported that the window of opportunity for hemispheric reorganization of speech differs for males and females. According to their findings, females had a relatively short window of reorganization—during the first year of life—while the window for language reorganization in males may extend to puberty. It has been suggested that sex differences in the outcome of early brain injury may be related to the relatively faster rate of maturation of the female brain during childhood (Strauss Wada Goldwater, 1992, Strauss Wada Hunter, 1992).

Studies that have considered factors that may be associated with atypical speech representation have tended to yield mixed findings. The method of establishing cerebral speech dominance has varied across studies, including procedures such as IAP, functional neuroimaging, cortical simulation, and dichotic listening tasks. Although some larger studies of speech dominance have been conducted with adult patients, samples of children have tended to be smaller, precluding extensive analysis of speech representation in children. The goal of the present study was to examine the relationship between various factors and the development of atypical speech representation in a relatively large sample of children who have undergone IAP.